Deeply Nested Code, Bugs and Distributed Systems

Oct 4th, 2017 1:47pm by Alex Williams

Also available on Apple Podcasts, Google Podcasts, Overcast, PlayerFM, Pocket Casts, Spotify, Stitcher, TuneIn

The nature of software is by default deeply nested, making bug hunting always a work of forensics. In distributed systems, there may be only a few lines of code but there can be a long trail of dependencies. It’s a chain of dependencies with bugs that might be anywhere.

To the surface comes a lot of issues not previously needed to address. “There may be tens of services involved — the bug could be anywhere,” said Gareth Rushgrove of Puppet who talked to us at the Software Circus event last month for this latest edition of The New Stack Makers podcast.

To connect the different components means all those bits have to work together. It’s why OpenAPI and gRPC are important for developers and engineers who are building these so-called “fragile systems” that work across a specific version of an API or multiple versions of a platform such as Kubernetes. All these loosely coupled pieces will have to work. It’s just a requirement if they are going to be used at all.

In this Edition:

1:04: Addressing the theme of deeply nested code.
2:59: Highlighting the problems with reliance on individual pieces of software when fixing bugs.
5:06: Building fragile systems using approaches such as with gRPC.
8:04: How finding the trail and sources of bugs is just the tip of the iceberg.
9:00: What Puppet is working on currently in the container space.

Dell EMC {Code} sponsored this podcast. Puppet is a sponsor of The New Stack.

Alex Williams is founder and publisher of The New Stack. He's a longtime technology journalist who did stints at TechCrunch, SiliconAngle and what is now known as ReadWrite. Alex has been a journalist since the late 1980s, starting at the...